Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021199
Kuo-Chu Chang, Z. Tian
A Bayesian network is a compact representation for probabilistic models and inference. They have been used successfully for multisensor fusion and situation assessment. It is well known that, in general, the inference algorithms to compute the exact posterior probability of the target state are either computationally infeasible for dense networks or impossible for mixed discrete continuous networks. In those cases, one approach is to compute the approximate results using simulation methods. This paper proposes efficient inference methods for those cases. The goal is not to compute the exact or approximate posterior probability of the target state, but to identify the top (most likely) ones in an efficient manner. The approach is to use intelligent simulation techniques where previous samples will be used to guide the future sampling strategy. By focusing the sampling on the "important" space, we are able to sort out the top candidates quickly. Simulation results are included to demonstrate the performances of the algorithms.
{"title":"Efficient inference for mixed Bayesian networks","authors":"Kuo-Chu Chang, Z. Tian","doi":"10.1109/ICIF.2002.1021199","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021199","url":null,"abstract":"A Bayesian network is a compact representation for probabilistic models and inference. They have been used successfully for multisensor fusion and situation assessment. It is well known that, in general, the inference algorithms to compute the exact posterior probability of the target state are either computationally infeasible for dense networks or impossible for mixed discrete continuous networks. In those cases, one approach is to compute the approximate results using simulation methods. This paper proposes efficient inference methods for those cases. The goal is not to compute the exact or approximate posterior probability of the target state, but to identify the top (most likely) ones in an efficient manner. The approach is to use intelligent simulation techniques where previous samples will be used to guide the future sampling strategy. By focusing the sampling on the \"important\" space, we are able to sort out the top candidates quickly. Simulation results are included to demonstrate the performances of the algorithms.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132388919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020948
Paolo Remagnino, Graeme A. Jones
The fusion of tracking and classification information in multi-camera surveillance environments will result in greater robustness, accuracy and temporal extent of interpretation of activity within the monitored scene. Crucial to such fusion is the recovery of the camera calibration which allows such information to be expressed in a common coordinate system. Rather than relying on the traditional time-consuming, labour-intensive and expert-dependent calibration procedures to recover the camera calibration, extensible plug-and-play surveillance components should employ simple learning calibration procedures by merely watching objects entering, passing through and leaving the monitored scene. In this work we present such a two stage calibration procedure. In the first stage, a linear model of the projected height of objects in the scene is used in conjunction with world knowledge about the average person height to recover the image-plane to local-ground-plane transformation of each camera. In the second stage, a Hough transform technique is used to recover the transformations between these local ground planes.
{"title":"Automated registration of surveillance data for multi-camera fusion","authors":"Paolo Remagnino, Graeme A. Jones","doi":"10.1109/ICIF.2002.1020948","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020948","url":null,"abstract":"The fusion of tracking and classification information in multi-camera surveillance environments will result in greater robustness, accuracy and temporal extent of interpretation of activity within the monitored scene. Crucial to such fusion is the recovery of the camera calibration which allows such information to be expressed in a common coordinate system. Rather than relying on the traditional time-consuming, labour-intensive and expert-dependent calibration procedures to recover the camera calibration, extensible plug-and-play surveillance components should employ simple learning calibration procedures by merely watching objects entering, passing through and leaving the monitored scene. In this work we present such a two stage calibration procedure. In the first stage, a linear model of the projected height of objects in the scene is used in conjunction with world knowledge about the average person height to recover the image-plane to local-ground-plane transformation of each camera. In the second stage, a Hough transform technique is used to recover the transformations between these local ground planes.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132385520","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020915
R. Kannan, S. Sarangi, S. Ray, S. S. Iyengar
We define the problem of maximal sensor integrity placement, that of locating sensors in n-dimensional grids with minimal vulnerability to enemy attack or sensor faults. We show a polynomial time algorithm for computing sensor integrity exists for sensors with unbounded ranges deployed over a 1D grid of points. We then present an integer linear programming (ILP) formulation for computing sensor integrity for unbounded range sensors over higher dimension grids.
{"title":"Minimal sensor integrity in sensor grids","authors":"R. Kannan, S. Sarangi, S. Ray, S. S. Iyengar","doi":"10.1109/ICIF.2002.1020915","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020915","url":null,"abstract":"We define the problem of maximal sensor integrity placement, that of locating sensors in n-dimensional grids with minimal vulnerability to enemy attack or sensor faults. We show a polynomial time algorithm for computing sensor integrity exists for sensors with unbounded ranges deployed over a 1D grid of points. We then present an integer linear programming (ILP) formulation for computing sensor integrity for unbounded range sensors over higher dimension grids.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"188 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133787399","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020964
W. Torrez, D. Bamber, I. Goodman, H. Nguyen
There is an obvious need to be able to integrate both linguistic-based and stochastic-based input information in data fusion. In particular, this need is critical in addressing problems of track association, including cyber-state intrusions. This paper treats this issue through a new insight into how three apparently distinct mathematical tools can be combined: "boolean relational event algebra" (BREA), "one point random set coverage representations of fuzzy sets" (OPRSC), and "complexity-reducing algorithm for near optimal fusion" (CRANOF).
{"title":"A new method for representing linguistic quantifications by random sets with applications to tracking and data fusion","authors":"W. Torrez, D. Bamber, I. Goodman, H. Nguyen","doi":"10.1109/ICIF.2002.1020964","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020964","url":null,"abstract":"There is an obvious need to be able to integrate both linguistic-based and stochastic-based input information in data fusion. In particular, this need is critical in addressing problems of track association, including cyber-state intrusions. This paper treats this issue through a new insight into how three apparently distinct mathematical tools can be combined: \"boolean relational event algebra\" (BREA), \"one point random set coverage representations of fuzzy sets\" (OPRSC), and \"complexity-reducing algorithm for near optimal fusion\" (CRANOF).","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122300403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021011
S. Populaire, Joëlle Blanc, Thierry Denœux, Philippe Ginestet
This paper presents a methodology for combining expert knowledge with information from statistical data, in classification and prediction problems. The method is based on (1) a case-based approach allowing to predict a quantity of interest from past cases in the form of a belief function, (2) Bayesian networks for modelling expert knowledge and (3) a tuning mechanism allowing to optimally discount information sources by optimizing a performance criterion. This methodology is applied to the prediction of chemical oxygen demand solubility in waste-water The approach is expected to be useful in situations where both small databases and partial expert knowledge are available.
{"title":"Fusion of expert knowledge with data using belief functions: a case study in waste-water treatment","authors":"S. Populaire, Joëlle Blanc, Thierry Denœux, Philippe Ginestet","doi":"10.1109/ICIF.2002.1021011","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021011","url":null,"abstract":"This paper presents a methodology for combining expert knowledge with information from statistical data, in classification and prediction problems. The method is based on (1) a case-based approach allowing to predict a quantity of interest from past cases in the form of a belief function, (2) Bayesian networks for modelling expert knowledge and (3) a tuning mechanism allowing to optimally discount information sources by optimizing a performance criterion. This methodology is applied to the prediction of chemical oxygen demand solubility in waste-water The approach is expected to be useful in situations where both small databases and partial expert knowledge are available.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114577808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020934
Richard Brooks, Nathan Orr, John Zachary, Christopher Griffin
Malicious network activity is rapidly increasing. To understand and engineer countermeasures to network attacks, we have developed cellular automata models of network flow dynamics and associated attacks. We describe the theoretical development of our model and compare it to existing models of network flow based on statistical physics. Using our model, we have found empirical evidence that a link exists between the behavior of a network and its entropy. This paper discusses potential extensions of this work to entropy-based intrusion detection systems (IDS).
{"title":"An interacting automata model for network protection","authors":"Richard Brooks, Nathan Orr, John Zachary, Christopher Griffin","doi":"10.1109/ICIF.2002.1020934","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020934","url":null,"abstract":"Malicious network activity is rapidly increasing. To understand and engineer countermeasures to network attacks, we have developed cellular automata models of network flow dynamics and associated attacks. We describe the theoretical development of our model and compare it to existing models of network flow based on statistical physics. Using our model, we have found empirical evidence that a link exists between the behavior of a network and its entropy. This paper discusses potential extensions of this work to entropy-based intrusion detection systems (IDS).","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116005266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020933
Daniel G. Schwartz, S. Stoecklin, E. Yilmaz
This paper reports progress on creating a case-based implementation of the well-known Snort intrusion detection system. Snort is a simple rule-based system that is known to suffer limitations, including both failure to detect certain kinds of intrusions and the frequent raising of false alarms. We believe that a case-based reasoning approach can provide a framework in which to incorporate more sophisticated artificial intelligence techniques that will help overcome some of these limitations. In addition, the present system is intended to apply more generally to other aspects of network security, as well as other domains related to protecting the nation's critical infrastructure. The system is being built using the modern software engineering technique known as "adaptive" or "reflective architectures," which will make it easily adaptable to other kinds of problem domain.
{"title":"A case-based approach to network intrusion detection","authors":"Daniel G. Schwartz, S. Stoecklin, E. Yilmaz","doi":"10.1109/ICIF.2002.1020933","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020933","url":null,"abstract":"This paper reports progress on creating a case-based implementation of the well-known Snort intrusion detection system. Snort is a simple rule-based system that is known to suffer limitations, including both failure to detect certain kinds of intrusions and the frequent raising of false alarms. We believe that a case-based reasoning approach can provide a framework in which to incorporate more sophisticated artificial intelligence techniques that will help overcome some of these limitations. In addition, the present system is intended to apply more generally to other aspects of network security, as well as other domains related to protecting the nation's critical infrastructure. The system is being built using the modern software engineering technique known as \"adaptive\" or \"reflective architectures,\" which will make it easily adaptable to other kinds of problem domain.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"708 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115125764","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020983
Wen-Ran Zhang
It is observed that Boolean logic is a unipolar logic defined in the unipolar space {0,1}. It is argued that a unipolar system cannot be directly used to represent and reason with the coexistence of bipolar truth. To circumvent the representational and reasoning limitations of unipolar systems, a 4-valued bipolar combinational logic BCL is introduced based on the ancient Chinese Yin-Yang philosophy. The new logic is defined in a strict bipolar space S = {- 1,0}/spl times/{0,1}, which is proved a generalization of Boolean logic and a fusion of two interactive unipolar subsystems. Bipolar tautologies including modus ponens are introduced for bipolar inference. The semantics of the new logic is established, justified, and compared with unipolar systems. Bipolar relations, bipolar transitivity, and polarized reflexivity are introduced. An O(n/sup 3/) algorithm is presented for bipolar transitive closure computation. In addition, the lair's case in the ancient paradox is redressed based on bipolar logic and bipolar relations.
{"title":"Bipolar logic and bipolar knowledge fusion","authors":"Wen-Ran Zhang","doi":"10.1109/ICIF.2002.1020983","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020983","url":null,"abstract":"It is observed that Boolean logic is a unipolar logic defined in the unipolar space {0,1}. It is argued that a unipolar system cannot be directly used to represent and reason with the coexistence of bipolar truth. To circumvent the representational and reasoning limitations of unipolar systems, a 4-valued bipolar combinational logic BCL is introduced based on the ancient Chinese Yin-Yang philosophy. The new logic is defined in a strict bipolar space S = {- 1,0}/spl times/{0,1}, which is proved a generalization of Boolean logic and a fusion of two interactive unipolar subsystems. Bipolar tautologies including modus ponens are introduced for bipolar inference. The semantics of the new logic is established, justified, and compared with unipolar systems. Bipolar relations, bipolar transitivity, and polarized reflexivity are introduced. An O(n/sup 3/) algorithm is presented for bipolar transitive closure computation. In addition, the lair's case in the ancient paradox is redressed based on bipolar logic and bipolar relations.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116530328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1021170
S. Wellington, J.D. Vincent
The Nadaraya-Watson (N-W) statistical estimator based on Haar kernels can be used to implement a fuser based on empirical data. Fuser design essentially consists of the following interrelated activities: select a set of n observations from a pool of p prior observations; select a value for the bandwidth. Optimal fuser design can therefore involve a very large search space. This paper proposes the use of a genetic algorithm (GA) to optimise the fuser design. The GA is used to evolve optimal values for the bandwidth and subset of observations used to implement the fuser. Indicative test results are provided. The N-W fuser is shown to perform better than the best single sensor. The GA provides better results than manual design optimisation, with the performance of the N-W fuser comparable to that achieved using a feedforward neural network.
{"title":"Design optimisation of the Nadaraya-Watson fuser using a genetic algorithm","authors":"S. Wellington, J.D. Vincent","doi":"10.1109/ICIF.2002.1021170","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1021170","url":null,"abstract":"The Nadaraya-Watson (N-W) statistical estimator based on Haar kernels can be used to implement a fuser based on empirical data. Fuser design essentially consists of the following interrelated activities: select a set of n observations from a pool of p prior observations; select a value for the bandwidth. Optimal fuser design can therefore involve a very large search space. This paper proposes the use of a genetic algorithm (GA) to optimise the fuser design. The GA is used to evolve optimal values for the bandwidth and subset of observations used to implement the fuser. Indicative test results are provided. The N-W fuser is shown to perform better than the best single sensor. The GA provides better results than manual design optimisation, with the performance of the N-W fuser comparable to that achieved using a feedforward neural network.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123534262","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2002-07-08DOI: 10.1109/ICIF.2002.1020910
Subhash Challa, Jonathan A. Legg
Fusing out-of-sequence information is a problem of growing importance due to an increased reliance on networked sensors embedded in complicated network architectures. The problem of fusing out-of-sequence measurements (OOSM) has received some attention in literature; however, most practical fusion systems, owing to compatibility with legacy sensors and limited communication bandwidth, send track information instead of raw measurements to the fusion node. Delays introduced by the network can result in the reception of out-of-sequence tracks (OOST). This paper considers the problem of fusing out-of-sequence measurements in general, and proposes an optimal Bayesian solution involving a joint probability density of current and past target states, referred to as augmented states. By representing tracks using equivalent measurements, the relationship between OOSM and OOST-based fusion is shown. The special case of Gaussian statistics is also addressed.
{"title":"Track-to-track fusion of out-of-sequence tracks","authors":"Subhash Challa, Jonathan A. Legg","doi":"10.1109/ICIF.2002.1020910","DOIUrl":"https://doi.org/10.1109/ICIF.2002.1020910","url":null,"abstract":"Fusing out-of-sequence information is a problem of growing importance due to an increased reliance on networked sensors embedded in complicated network architectures. The problem of fusing out-of-sequence measurements (OOSM) has received some attention in literature; however, most practical fusion systems, owing to compatibility with legacy sensors and limited communication bandwidth, send track information instead of raw measurements to the fusion node. Delays introduced by the network can result in the reception of out-of-sequence tracks (OOST). This paper considers the problem of fusing out-of-sequence measurements in general, and proposes an optimal Bayesian solution involving a joint probability density of current and past target states, referred to as augmented states. By representing tracks using equivalent measurements, the relationship between OOSM and OOST-based fusion is shown. The special case of Gaussian statistics is also addressed.","PeriodicalId":399150,"journal":{"name":"Proceedings of the Fifth International Conference on Information Fusion. FUSION 2002. (IEEE Cat.No.02EX5997)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2002-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124668903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：